关键词: 固沙植被, 土壤物理因子, 变异函数, 空间异质性

Abstract: With the development of desertification in the world getting more and more serious at present, the exploring of the methods reversing desertification is especially important. The establishment of sand-fixing artificial vegetation in Shapotou area is a good example of controlling desert. The achievement and the pattern of reversing desertification have been recognized by many scientists in the world. The artificial sand-fixing vegetation in Shapotou area was primarily established in 1956. With the succession of the artificial vegetation for more than 40 years, an artificial-natural vegetation ecosystem has been formed. Many former scholars studied this ecosystem in the aspects of plants, soil physical and chemical properties, soil moisture, soil microbiologic crusts and climatology. However, there is little information on spatial variability and scales of soil physical factors in this vegetation area. The present study aims to address the general need for basic information about soil physical factor heterogeneity and to offer important reference information for the further relationship study between soil heterogeneity and vegetation patterns by using principle and methods of geostatistics. In Shapotou artificial vegetation area, the soil moisture, bulk density in 0~15 cm layer are less than that in 15~30 cm layer; the capillary moisture capacity and porosity are reversed, whereas the variation coefficients of 0~15 cm layer are more than that of 15~30 cm layer. Results of semi-variance analysis show that there are high degrees of spatial heterogeneity in soil moisture, bulk density, capillary moisture and porosity in 0~15cm layer. The scale of spatial heterogeneity is 13.8~28.2 m. The spatial heterogeneity of autocorrelations in total spatial heterogeneity are 83.5%~99.9%, which are far more than the spatial heterogeneity of random in total spatial heterogeneity. There is no spatial heterogeneity for each soil physical factor in 15~30 cm layer. The semi-variance function of soil moisture is stochastic model, and the semi-variance function of bulk density, capillary moisture capacity and porosity are linear modals. On the basis of Kriging interpolation, the distribution maps are produced, from which the distributions of each soil physical factors can be seen clearly. The relationship between spatial heterogeneity and vegetation is discussed.

Key words: sand-fixing vegetation, soil physical factors, semi-variance, heterogeneity